6 ms·
I work in media (for Mashable, specifically, though not in an editorial capacity), and I'm puzzled as to how people are getting native advertising so wrong. I'm
by cheald 12y ago
I work in media (for Mashable, specifically, though not in an editorial capacity), and I'm puzzled as to how people are getting native advertising so wrong. I'm aware that my paycheck is linked to this sort of thing, so full biases disclosed, but I really do think that there's a misunderstanding of what it actually is. There's this tendency to conflate "someone paid money to have their name attached to this article" with "someone paid to have a good article written about them", and they're wildly different.
(I should disclaim that these are my personal opinions as someone in the industry, and I don't claim to speak on behalf of my employer)
"Native advertising" like The Atlantic's Scientology advertorial is unequivocally bad. It's sneaky, it's underhanded, it purports to be unbiased reporting when it's anything but. However, that is really the exception rather than the rule. (It's worth noting that this has been the status quo in print magazines for quite some time; brands provide 1- or 2-page ads which are presented to look like an article that belongs in the magazine, with a tiny bit of "This is an advertisement" text stuffed in a corner. I'll scan a few if folks are interested. They're massively worse than the sort of native advertising under discussion here.)
"Good" native advertising is the practice of letting brands pay to attach their names to thematically-relevant articles, whose content is not bought, directed by, or advertising the brand who wishes to advertise on it. This pays really well for media outlets, converts well for advertisers, and results in less obtrusive, annoying advertising for readers. For example, from the FTC workshop on native advertising:
> [An example] is American Express, who came to us looking to reach female small business owners. So what we created on Mashable was a site called -- sorry, a content series, including videos and articles and info-graphics called "The Female Founders Series" where we profiled female entrepreneurs in technology, profiles and videos and vignettes, that we published on Mashable that were presented by American Express.
(http://www.ftc.gov/system/files/documents/public_events/171321/final_transcript_1.pdf http://www.ftc.gov/system/files/documents/public_events/1713... page 43)
In this case, profiling female startup founders is directly in Mashable's wheelhouse; it is relevant to our audience (which is socially and technologically savvy, young, and a majority of whom are female), and it's relevant to our advertiser. At no point are Amex's services pushed or touted or even talked about. Amex doesn't get to write the articles or have any editorial control over them. Our KPIs are typically engagement, so in order to deliver on our end of the deal, we have to provide content that people enjoy and want to share. The sponsorship is clearly disclosed in the interest of transparency, but at no point does the sponsor get to inject their brand, agenda, or marketing fluff into the actual content; it is always adjacent to it in the form of "sponsored by" highlights and traditional ad units.
The key difference here is that Amex can say "We want our name to be in front of people for every page view of this series of content", rather than "We are going to do a traditional ad buy which will be demographically targeted to US females 19-30 years old, which may or may not end up associated with this series of content". They can't say "Write about this founder, and tell the story of how Amex made her business succeed"; we flat out will not do that.
I'm of the opinion that this kind of advertising is actually better for all involved, as long as editorial independence is maintained. All media is sponsored at some level - advertising drives the entire industry - so if the benchmark for "good" content is "content that advertisers don't have any stake in", then...well, good luck. Advertisements that advertisers push through otherwise-respected media outlets in the guise of articles written by the outlet from a journalistic standpoint are bad, but they are the tiny minority of native advertising.
- opendais 12y agoI don't think you get the conflict of interest issue, like, at all. Your example is sort of okay. However, its really easy to find examples that aren't okay from your own site. 1) http://mashable.com/2013/04/11/social-ad-revenue/ http://mashable.com/2013/04/11/social-ad-revenue/ Ad revenue article sponsored by a marketing firm that sells a product in relation to that. 2) http://mashable.com/2011/11/18/hubspot-sponsored-post-1/ http://mashable.com/2011/11/18/hubspot-sponsored-post-1/ This is basically plugging what Hubspot is selling. 3) http://mashable.com/2010/12/15/sponsored-technology-500-list/ http://mashable.com/2010/12/15/sponsored-technology-500-list... Here is another one that basically plugs itself.
- cheald 12y agoI do get it; it's something that we're acutely aware of, and there's a lot of effort being put into improving the integrity and transparency of the whole affair. The first article is syndicated from ClickZ - they're one of our syndication partners, but I'm not sure what the criteria for syndication are. It's not actually clear that there was any sponsorship by BIA/Kelsey on it (I think that's who you're talking about?) The other two examples are much worse, granted, but it's worth noting that the other two examples are almost 3+ years old at this point; we ran with that for a bit while trying to figure out native advertising, but the company has moved on from that kind of advertising. You'll also notice the giant [SPONSORED] in the headline, slug, and highly visible advertising disclaimers at the top and bottom of the post. There is no ambiguity about their "bought" status.
- opendais 12y agoAlright. I'll break down the Clickz example since it seems the most relevant to your "current practices" at Mashable. I honestly just googled "sponsored posts Mashable" and went down this list: http://mashable.com/category/sponsored-post/ http://mashable.com/category/sponsored-post/ I didn't bother looking at the dates. Clickz: http://www.clickzlive.com/ http://www.clickzlive.com/ http://www.clickzacademy.com/training.php http://www.clickzacademy.com/training.php They sell information-oriented products for marketing and are reporting on the size of such markets. That is a huge, massive conflict of interest. It is like me saying "Look at this huge market over there, here let me help you learn how to tap into it...for $199" and then publishing it as if it was a newsworthy article. So you have a company that is pushing a product while simultaneously publishing articles related to the value of that product without editorial oversight. Does that not seem like a conflict of interest to you? Simply blindly syndicating that sort of thing is the same as native advertising imo.